Based on patch by Vittorio Giovara <vittorio.giovara@gmail.com> from 696141e898

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Reported-by: Vittorio Giovara <vittorio.giovara@gmail.com>
Inspired by discussion with Kieran Kunhya <kierank@obe.tv>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Based-on: 2e1704059a by Kieran Kunhya

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

This permits replacing it by a optimized implementation
Based-on / Idea-from: 2e1704059a by Kieran Kunhya

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Idea from 2e1704059a from Kieran Kunhya

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Aleksey Vasenev <margtu-fivt@ya.ru>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>

The purpose of this filter mode is to allow interlaced content to
display properly in interlaced video modes, as described in
http://forum.xbmc.org/showthread.php?tid=81834 and
https://github.com/mpv-player/mpv/issues/624#issuecomment-37685195 . The
filter doubles the video frame rate, but does not work properly because:
(1) it does not set the properties of the output stream to indicate the
doubled frame rate, and
(2) it does not set an appropriate PTS on the extra frames.
The attached patch fixes these problems by settling these values the
same way they are set in vf_yadif mode 1 (field) which also doubles the
frame rate.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Inspired by: 3a16ec19d2

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Paul B Mahol <onemda@gmail.com>

Signed-off-by: Paul B Mahol <onemda@gmail.com>

Signed-off-by: Paul B Mahol <onemda@gmail.com>

This should fix several issues with odd dimensions inputs.

lut, vflip, pad and crop video filters also need to be checked for such
issues. It's possible sws is also affected.

Fixes green bottom line

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Paul B Mahol <onemda@gmail.com>

Fixes CID966642

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Stefano Sabatini <stefasab@gmail.com>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Simplify code, and provide introspection through the AVOption system.

Also add missing NULL checks.

This is following 568c70e79e.

Fixes CID703717
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

That interface is deprecated, and apparently useless.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Increase performance (slightly).

If not set to NULL, the reference is freed by ff_end_frame(), and later
accessed in end_frame() by the filter code, causing a crash.

Reduce log clutter, consistent with 1a49a169eb.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Fix signature mismatch warnings.

Signed-off-by: Paul B Mahol <onemda@gmail.com>

Also deprecate the use of numerical values.

This is useful for marking progressive video processed by the filter as
interlaced, avoiding the interlaced flag to switch back and forth at each
frame.

This new mode is useful for generating frames for interlaced video
displays. Typically interlaced video displays have no form of field
synchronisation. This new mode guarantees correct field order without
any requirement for field synchronisation.

Signed-off-by: Stefano Sabatini <stefasab@gmail.com>

Allow creating interlaced bottom field first video.

Port MPlayer tinterlace filter from MPlayer, with some ideas taken
from the FFmbc/libavfilter port, with the following main differences:

* added support for full-scale YUVJ formats
* added support for YUVA420P
* request_frame() on the filter is forced to return a frame
* some code factorization (related to the copy_picture_fields() function)
* fixed black padding values for mode 3